There has long been evidence that RA, the reactive arthritides (including AIDS-associated Reiter's syndrome) and related disorders all involve both genetic and infectious etiologic factors. Recently, increased attention is focusing on the role of the suspected infectious etiologic components of these diseases. In the past, an immense effort has been made to detect such agents in these disorders using chiefly culture and immunologic methods, without notable success. These approaches suffer from the need for prior knowledge about the growth requirements or immunology of the organisms being sought and, therefore, have been narrow in scope. Thus, the many negative results obtained cannot easily be interpreted. To justify another such effort, it is essential to overcome these limitations. The present approach is believed largely to achieve this. It employs advances in molecular biology and knowledge of ribosomal base sequences that have become available in the past 2-3 years and can detect most or all eubacteria, a group which contains essentially all prokaryotes of medical significance, including Chlamydiae and Rickettsiae, in clinical specimens, with a sensitivity equivalent to the DNA content of 10 E. coli organisms. It employs synthetic oligonucleotides with broad eubacterial homology to amplify microbial DNA extracted from clinical specimens by the polymerase chain reaction (PC). Detected organisms can be characterized sufficiently well to help distinguish contaminants from true pathogens. In addition to fresh specimens, formalin-fixed, embedded pathology specimens can also be studied, thus allowing examination of otherwise rare tissues. Studies of known septic arthritis demonstrated the feasibility of this method. Most interestingly, preliminary data on RA have provided evidence for the presence of eubacteria in synovial fluid cells, with negative controls from gout and osteoarthritis. By extending and confirming these results, exploring related disorders including AIDS-associated Reiter's syndrome, and elucidating the etiologic significance fo detected organisms, information central to the understanding and management of these diseases may be obtained.